AWEAR Studies: Data Checker

This notebook was written for the AWEAR Case Studies as part of a PhD project on Engineering Systems Design in Healthcare at the Engineering Systems Division, DTU Management, in collaboration with Rigshospitalet-Glostrup.

The purpose of this notebook is to provide regular reports on the data being recorded from study participants, to check that it is being recorded and looks as expected. The R script imports, tests and plots data from a specified participant and timeframe, as shown in the code and output below.

Report subject and period:

This report is for user  P10JL for the time period from  2018-03-19  to  2018-03-30

Import and restructure data:

Overview of available csv files and their current use status:

File	Description	Status
“activity.csv”	coded activity types	in use
“battery.csv”	charging log	in use
“bluetooth.csv”	-	not in use
“calllog.csv”	log of phonecalls	to be added
“experience_sampling.csv”	answers to daily self-reports	in use
“hardware_info.csv”	-	not in use
“location.csv”	GPS data from watch and phone	in use
“screen.csv”	screen on/off transitions (phone)	in use
“sms.csv”	sms log	to be added
“step_count.csv”	step counts from watch and phone	in use
“wearable.csv”	-	no longer in use
“wifi.csv”	-	not in use

The output below confirms which files have been imported, followed by processing steps.

activity file imported
battery file imported
calllog file imported
experience_sampling file imported
location file imported
screen file imported
sms file imported
step_count file imported

 *** 
 Processing file containing:  activity confidence 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch
Complete: column 'label' created describing activity based on activity code 
Complete: activity data reduced from  4397  to  2618  rows by keeping only the max-confidence activities for each unique timestamp 

 *** 
 Processing file containing:  plugged status level 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch
Complete: column 'mode' created describing charging status based on 'plugged' variable 

 *** 
 Processing file containing:  name number type duration 

User ID is not in dataset, returning NULL 

 *** 
 Processing file containing:  question_id answer 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch

 *** 
 Processing file containing:  lat lon accuracy provider 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch

 *** 
 Processing file containing:  screen_on 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch

 *** 
 Processing file containing:  address body type read 

User ID is not in dataset, returning NULL 

 *** 
 Processing file containing:  step_count 

Complete: Data extracted for specified user 
Complete: Data extracted for period of interest
Complete: funf_version converted to dsource column indicating whether reading is from phone or watch

Quality check: data source

Check whether data coming from watch, phone or both (also gives date of last reading).

Data in  activity comes from  phone 
Last reading in activity is on [1] "2018-03-28 07:31:49 CEST" 

Data in  battery comes from  phone 
Last reading in battery is on [1] "2018-03-28 07:31:49 CEST" 

Data in  experience_sampling comes from  phone 
Last reading in experience_sampling is on [1] "2018-03-27 14:30:54 CEST" 

Data in  location comes from  phone watch 
Last reading in location is on [1] "2018-03-28 07:36:35 CEST" 

Data in  screen comes from  phone 
Last reading in screen is on [1] "2018-03-28 07:36:42 CEST" 

Data in  step_count comes from  phone watch 
Last reading in step_count is on [1] "2018-03-28 07:37:00 CEST" 

Visualise data:

to do

• resolve warnings
• customise colours
• axes with dates: orders not in time, january after feb (ordered by day not taking month into account)

minimal value for n is 3, returning requested palette with 3 different levels
minimal value for n is 3, returning requested palette with 3 different levels

The shape palette can deal with a maximum of 6 discrete values because more than 6
becomes difficult to discriminate; you have 11. Consider specifying shapes manually if
you must have them.The shape palette can deal with a maximum of 6 discrete values because more than 6
becomes difficult to discriminate; you have 11. Consider specifying shapes manually if
you must have them.